My personal experience of 🦀Crab-ification of E-commerce product design

“real life example by two different companies”

So, what is Crab-ification?

Info

Carcinization is a form of convergent evolution in which non-crab crustaceans evolve a crab-like body plan.

Convergent evolution is the independent evolution of similar features in species of different periods or epochs in time.

In morphology, analogous traits arise when different species live in similar ways and/or a similar environment.

Convergent Evolution in Product Development

Convergent evolution in product development occurs when different companies independently create products or features that are similar in function or design, despite starting from different initial conditions or development paths. This similarity arises because these companies are addressing comparable market needs, user behaviors, or technological constraints, leading them to converge on analogous solutions.

Background of my personal experience

Good Company

I was working as one of the senior product managers at a leading audio streaming company (good company), which is trying to build an internal commerce platform to expand the core capabilities of its existing payment infrastructure and accommodate more “shopping behaviors” to support its internal business verticals besides subscription business, for example, during my tenure there, it expanded with following business verticals:

Podcast business, paywall, paid episodes
Audiobooks, either purchase as A La Carte, Top-ups
Concert Tickets resell
etc

Bad Company

I happened to got a contract position to making money at a top consumer brand company, their VP approached me for a similar opportunity to help them build or improve their “Checkout” experience. It turned out, checkout at Bad Company is a loosen definition not mapping into a specific portion of the user journey.

Side by Side Comparison

Highlevel User Journey

Intent to Buy(Upper Funnel): product discovery, different types of products, brands will lead to different product discovery processes
Intent to Pay(Lower Funnel): once customers finalize the items in their cart, their mindset are switch to focusing on make the payment, as the platform, a smooth and repeatable checkout process is required, which should eliminate any possible confusions and traffic leaks that brings users back to Upper funnel.

Productstatement

A typical consumer facing commerce platform, is to accommodate user’s purchase journey: discovery of product then decide to pay.

We can use a Intent Model to look at this problem, promote more traffic from Intent to Buy → Intent to Pay

Conversion

is the ratio between Intent to Pay 💳 and Intent to Buy 🛒

🛍️ = 💳 / 🛒

****	Good Company	Bad Company	Personal Thoughts
Naming	Unified Checkout💳 A lot thoughts behind the naming, understanding, what part of users experience could be “Unified”, what part of users experience couldn’t and shouldn’t be “unified”.	Unified Purchase👀🛒💳 Try to unify entire “purchase” process, which is a mission impossible.	How you name things really matters, which reflects your understanding of a problem you solve, what’s more, naming things impact human behaviors. Upper Funnel should enjoy the diversity, make product discovery process interesting, which Lower Funnel is to make it ""boring and stable"", make ""payment process"" dumb. Think about ""Express checkout"" etc.
User Research Model	An Intent Model of Buy vs Pay, understand for each stage, how customers think, why they think, and based on that, what features to build to accommodate that type of thinking, it’s a funnel, but it is more than a funnel.	No such thing as User Research Model, generic concepts like top funnel and bottom funnel are circulated within the company, but no one ever gives a clear definition	A user research model is to establish the fundamentals about how team view a problem and how to solve it. It lays a common platform for product team, ops team, engineering teams to building things that reflect the actual business, user journey as true to reality as possible.
Tech Stack	Different business verticals were acquired and being integrated into the same tech stack, with not just code reconciliation but also naming of variables, architecture designs and core capabilities mapping. https://payments.spotify.com/checkout/spotify/9fbb6311-de7e-31b2-9c8d-64c5db0a7bd2/?country=US&market=us&product=default-full-price-premium Once you Entering “checkout session”, everything is on one page! [Screenshot 01]	Two processes are ongoing at the same time, moving from legacy code base to new code base, while each commerce experiences with it’s own source of requirements, the “unification” is really happening only on pure visual part instead of “experience”. e.g.: https://www.esteelauder.com/checkout/viewcart.tmpl This URL reflects that engineers think CART is part of checkout process Which indicates a wrong “unification”	Like how Convergent Evolution works, the response to external environment is the underlying driven force, different types of understanding on how to response the external environment will leads similar features at different places with different purposes. e.g.: At Bad Company, https://www.esteelauder.com/checkout/samples.tmpl After users clicked “Checkout” Button, a sample person still inserted to ask customer to switch back to their previous intent to buy NOT to pay. [Screen Recording 01]. Thus, “Checkout” just means continue shopping here, instead of means the “starting point” or “user signal” indicates that customers are ready to pay, which leads potential drop off.
Success Measures	Given clear User Research Model depicts the right user journey, metrics and goals are set accordingly for each stage of the experience. e.g.: Overall conversion rate, Upper and Lower Funnel Conversion Rate etc.	Only one Overall Conversion Rate is set, even used for Experiments, which often causing problems to analyze the bottleneck of performances.	Other difference, here, the Bad company is main a web environment, unlike the mobile app environment, users are able to bypass steps by input page URL, thus, capturing user sent signal could be tricky, for example, just because a user landed on checkout related pages, if we use page view, that’s not a good signal to approximate users are ready to pay, beside, due the the design on Bad Company’s website, user’s direct interaction on “CHECKOUT” button might also mis leading, due to the loosen definition, like the sample example above, it simply could mean “CONTINUE SHOPPING”.
Result	Good design could be even ensure this company’s success under the restriction from Apply pay, the good company does not surrender to Apple to pay a 30% fee, but with its good UI/UX design, they offset the amount of effort asking users to jump out of app to finish a quick checkout process on Website.	See, so many back and forth and traffic leaks when users are entering checkout stage, low overall conversion rate.	But Bad Company still able to survive well, given the underlying business model is built upon “human nature”